hc = hail.HailContext(log="/hail.log")
logger.info("\n==> import vds: " + input_vds_path)
vds = hc.read(input_vds_path)
parallel_computed_annotation_exprs = [
"va.variantId = %s" % get_expr_for_variant_id(),
"va.contig = %s" % get_expr_for_contig(),
"va.start = %s" % get_expr_for_start_pos(),
"va.pos = %s" % get_expr_for_start_pos(),
"va.end = %s" % get_expr_for_end_pos(),
"va.ref = %s" % get_expr_for_ref_allele(),
"va.alt = %s" % get_expr_for_alt_allele(),
"va.xpos = %s" % get_expr_for_xpos(pos_field="start"),
"va.xstart = %s" % get_expr_for_xpos(pos_field="start"),
]
serial_computed_annotation_exprs = [
"va.xstop = %s" % get_expr_for_xpos(field_prefix="va.", pos_field="end"),
]
vds = vds.annotate_variants_expr(parallel_computed_annotation_exprs)
vds = vds.annotate_variants_expr(serial_computed_annotation_exprs)
# apply schema to dataset
INPUT_SCHEMA = {
"top_level_fields": """
variantId: String,
originalAltAlleles: Set[String],
ReadPosRankSum: Double,
SOR: Double,
VQSLOD: Double,
culprit: String,
AC_Hom: Array[Int],
AC_Het: Array[Int],
AC_Hemi: Array[Int],
"""
}
vds_computed_annotations_exprs = [
"va.chrom = %s" % get_expr_for_contig(),
"va.pos = %s" % get_expr_for_start_pos(),
"va.ref = %s" % get_expr_for_ref_allele(),
"va.alt = %s" % get_expr_for_alt_allele(),
"va.xpos = %s" % get_expr_for_xpos(),
"va.variantId = %s" % get_expr_for_variant_id(),
"va.originalAltAlleles = %s" % get_expr_for_orig_alt_alleles_set(),
"va.geneIds = %s" % get_expr_for_vep_gene_ids_set(),
"va.transcriptIds = %s" % get_expr_for_vep_transcript_ids_set(),
"va.transcriptConsequenceTerms = %s" % get_expr_for_vep_consequence_terms_set(),
"va.sortedTranscriptConsequences = %s" % get_expr_for_vep_sorted_transcript_consequences_array(),
"va.mainTranscript = %s" % get_expr_for_worst_transcript_consequence_annotations_struct("va.sortedTranscriptConsequences"),
"va.sortedTranscriptConsequences = json(va.sortedTranscriptConsequences)"
]
print("======== Exomes: KT Schema ========")
for expr in vds_computed_annotations_exprs:
vds = vds.annotate_variants_expr(expr)
kt_variant_expr = convert_vds_schema_string_to_vds_make_table_arg(split_multi=False, **SCHIZOPHRENIA_SCHEMA)
pprint(kt_rare_variants.schema)
ES_HOST_IP = '10.4.0.13'
ES_HOST_PORT = 9200
print("======== Export to elasticsearch ======")
es = ElasticsearchClient(
host=ES_HOST_IP,
port=ES_HOST_PORT,
)
annotation_expressions = [
'variant_id = %s' % get_expr_for_variant_id(),
'chrom = %s' % get_expr_for_contig(),
'pos = %s' % get_expr_for_start_pos(),
"xpos = %s" % get_expr_for_xpos(field_prefix="", pos_field="pos"),
]
for expression in annotation_expressions:
kt_rare_variants = kt_rare_variants.annotate(expression)
kt_rare_variants = kt_rare_variants.drop(['v'])
kt_annotations = kt_annotations.annotate('variantId = %s' % get_expr_for_variant_id()).drop(['v'])
kt_rare_variants = kt_rare_variants.key_by('variantId').join(kt_annotations.key_by('variantId'))
pprint(kt_rare_variants.schema)
es.export_kt_to_elasticsearch(
kt_rare_variants,
"""
vds = vds.annotate_variants_expr("va.originalAltAlleles=%s" % get_expr_for_orig_alt_alleles_set())
vds = add_mpc_to_vds(hc, vds, args.genome_version, root="va.info", info_fields=MPC_INFO_FIELDS)
pprint(vds.variant_schema)
for expr in vds_computed_annotations_exprs:
vds = vds.annotate_variants_expr(expr)
kt_variant_expr = convert_vds_schema_string_to_vds_make_table_arg(**GNOMAD_SCHEMA)
# print kt_variant_expr
kt = vds.make_table(kt_variant_expr, [])
# pprint(kt.schema)
kt = kt.annotate("pos = start")
kt = kt.annotate("stop = %s" % get_expr_for_end_pos(field_prefix="", pos_field="start", ref_field="ref"))
kt = kt.annotate("xpos = %s" % get_expr_for_xpos(field_prefix="", pos_field="start"))
kt = kt.annotate("xstart = %s" % get_expr_for_xpos(field_prefix="", pos_field="start"))
kt = kt.annotate("xstop = %s" % get_expr_for_xpos(field_prefix="", pos_field="stop"))
# flatten and prune mainTranscript
transcript_annotations_to_keep = [
"amino_acids",
"biotype",
"canonical",
"cdna_start",
"cdna_end",
"codons",
#"distance",
"domains",
"exon",
"gene_id",
